{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "4c50c508",
   "metadata": {},
   "source": [
    "# Supplementary and Corraborative analyses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "589ce56c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys, os\n",
    "import glob\n",
    "import json\n",
    "import pprint\n",
    "import pandas as pd \n",
    "pd.options.mode.chained_assignment = None\n",
    "import seaborn as sns\n",
    "import numpy as np \n",
    "from scipy.stats import pearsonr\n",
    "\n",
    "\n",
    "from utils import *\n",
    "from mediation_estimands import * \n",
    "from filter import go_join_filter\n",
    "\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5cb1e124",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "if not os.path.exists('./figs'):\n",
    "    os.makedirs('./figs')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "41a87457",
   "metadata": {},
   "source": [
    "### Load data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e4659a21",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'chunk_path': 'data/chunks1.0/',\n",
      " 'end_year': 2020,\n",
      " 'exclude_adv_first_utt': False,\n",
      " 'exclude_backchannel': False,\n",
      " 'final_df_path': 'data/df_final.csv',\n",
      " 'include_fem_issue': False,\n",
      " 'min_num_chunks_per_just': 1000,\n",
      " 'num_bootstrap_samples': 1000,\n",
      " 'prev_utt_path': 'data/prev_utt_1.0/',\n",
      " 'start_year': 1982}\n"
     ]
    }
   ],
   "source": [
    "config = load_config() #Config file with pipeline decisions \n",
    "pprint.pprint(config)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "02fd8eaf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'data/chunks1.0/'"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "config['chunk_path']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "fdcf52b2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "num_exclude_adv_first_utt=0\n",
      "num chunks = len(df)= 75025\n"
     ]
    }
   ],
   "source": [
    "df_raw = load_chunks_df(config)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "016caa84",
   "metadata": {},
   "source": [
    "### Table A1 (Appendix):  Conditional means of interruption, Y given whether the case topic is about gender issues or not (C)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5a8644ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "original dataset num = 75025\n",
      "dataset w/ {0, 1} ideology mathces num = 68301\n",
      "Number of justices with >1000 chunks 15\n",
      "\t ['Anthony M. Kennedy', 'Antonin Scalia', 'Byron R. White', 'David H. Souter', 'Elena Kagan', 'John G. Roberts Jr.', 'John Paul Stevens', 'Ruth Bader Ginsburg', 'Samuel A. Alito Jr.', \"Sandra Day O'Connor\", 'Sonia Sotomayor', 'Stephen G. Breyer', 'Thurgood Marshall', 'Warren E. Burger', 'William H. Rehnquist']\n",
      "before justice filter, num chunks = 68301\n",
      "after justice filter, num chunks = 65756\n",
      "Saved final df to -> data/df_final.csv\n"
     ]
    }
   ],
   "source": [
    "# Include all data (with or without female issues)\n",
    "config[\"include_fem_issue\"] = True\n",
    "df_fem = go_join_filter(df_raw, config)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "31c87de1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "original dataset num = 75025\n",
      "dataset w/ {0, 1} ideology mathces num = 68301\n",
      "Number of justices with >1000 chunks 15\n",
      "\t ['Anthony M. Kennedy', 'Antonin Scalia', 'Byron R. White', 'David H. Souter', 'Elena Kagan', 'John G. Roberts Jr.', 'John Paul Stevens', 'Ruth Bader Ginsburg', 'Samuel A. Alito Jr.', \"Sandra Day O'Connor\", 'Sonia Sotomayor', 'Stephen G. Breyer', 'Thurgood Marshall', 'Warren E. Burger', 'William H. Rehnquist']\n",
      "before justice filter, num chunks = 68301\n",
      "after justice filter, num chunks = 65756\n",
      "Exluded female issues, num chunks 64164\n",
      "Saved final df to -> data/df_final.csv\n"
     ]
    }
   ],
   "source": [
    "config[\"include_fem_issue\"] = False\n",
    "df_final = go_join_filter(df_raw, config)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "56724027",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df_fem.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d2d6ec94",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "E[Y|C = Other issue,T = M] 10.9\n",
      "E[Y|C = Gender issue, T = M] 12.47\n",
      "E[Y|C = Other issue, T = F] 12.35\n",
      "E[Y|C = Gender issue, T = F] 10.82\n"
     ]
    }
   ],
   "source": [
    "e1 = df[(df['advocate_gender']=='M') & (df['female_issue']==0)]['adv_interruption_rate'].mean() \n",
    "print(\"E[Y|C = Other issue,T = M]\", np.round(e1, 2))\n",
    "e2 = df[(df['advocate_gender']=='M') & (df['female_issue']==1)]['adv_interruption_rate'].mean() \n",
    "print(\"E[Y|C = Gender issue, T = M]\", np.round(e2, 2))\n",
    "e3 = df[(df['advocate_gender']=='F') & (df['female_issue']==0)]['adv_interruption_rate'].mean() \n",
    "print(\"E[Y|C = Other issue, T = F]\", np.round(e3, 2))\n",
    "e4 = df[(df['advocate_gender']=='F') & (df['female_issue']==1)]['adv_interruption_rate'].mean() \n",
    "print(\"E[Y|C = Gender issue, T = F]\", np.round(e4, 2))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "197cfe5f",
   "metadata": {},
   "source": [
    "### Figure A1: Heatedness "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "8493adab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded final df from  data/df_final.csv\n",
      "Number of rows= 64164\n"
     ]
    }
   ],
   "source": [
    "df = load_final_df(config) # Dataframe after filtering to filters in config.yaml \n",
    "                           # (e.g., filtering to justices with the min num chunks) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "28655c02",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x648 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set()\n",
    "rows=2\n",
    "cols=2\n",
    "fig, axes = plt.subplots(rows, cols, figsize=(12,9))\n",
    "\n",
    "df_f_f = df[(df.advocate_gender=='F') & (df.justice_gender=='F')]\n",
    "df_f_f = df_f_f[['adv_interruption_rate', 'justice_interruption_rate']]\n",
    "\n",
    "df_f_f = df_f_f[(df_f_f.adv_interruption_rate!=0) & (df_f_f.justice_interruption_rate!=0)]\n",
    "df_f_f['adv_interruption_rate'] = df_f_f.apply(lambda x: np.log(x['adv_interruption_rate']), axis=1)\n",
    "df_f_f['justice_interruption_rate'] = df_f_f.apply(lambda x: np.log(x['justice_interruption_rate']), axis=1)\n",
    "\n",
    "corr, _ = pearsonr(df_f_f.adv_interruption_rate, df_f_f.justice_interruption_rate)\n",
    "corr = np.round(corr, 2)\n",
    "#print('f-f', corr)\n",
    "axes[0][0].scatter(df_f_f.adv_interruption_rate, df_f_f.justice_interruption_rate, s=2)\n",
    "axes[0][0].set_title('female advocate, female justice', fontsize=12)\n",
    "axes[0][0].axes.text(0.8,0.9, r'$\\rho$='+str(corr), horizontalalignment='left',\\\n",
    "                                 verticalalignment='center', transform=axes[0][0].transAxes,\\\n",
    "                                 fontsize=12)\n",
    "\n",
    "\n",
    "df_f_f = df[(df.advocate_gender=='F') & (df.justice_gender=='M')]\n",
    "df_f_f = df_f_f[['adv_interruption_rate', 'justice_interruption_rate']]\n",
    "\n",
    "df_f_f = df_f_f[(df_f_f.adv_interruption_rate!=0) & (df_f_f.justice_interruption_rate!=0)]\n",
    "df_f_f['adv_interruption_rate'] = df_f_f.apply(lambda x: np.log(x['adv_interruption_rate']), axis=1)\n",
    "df_f_f['justice_interruption_rate'] = df_f_f.apply(lambda x: np.log(x['justice_interruption_rate']), axis=1)\n",
    "\n",
    "corr, _ = pearsonr(df_f_f.adv_interruption_rate, df_f_f.justice_interruption_rate)\n",
    "corr = np.round(corr, 2)\n",
    "#print('f-m', corr)\n",
    "axes[0][1].scatter(df_f_f.adv_interruption_rate, df_f_f.justice_interruption_rate, s=2)\n",
    "axes[0][1].set_title('female advocate, male justice', fontsize=12)\n",
    "axes[0][1].axes.text(2,0.9, r'$\\rho$='+str(corr), horizontalalignment='left',\\\n",
    "                                 verticalalignment='center', transform=axes[0][0].transAxes,\\\n",
    "                                 fontsize=12)\n",
    "\n",
    "\n",
    "df_f_f = df[(df.advocate_gender=='M') & (df.justice_gender=='F')]\n",
    "df_f_f = df_f_f[['adv_interruption_rate', 'justice_interruption_rate']]\n",
    "\n",
    "df_f_f = df_f_f[(df_f_f.adv_interruption_rate!=0) & (df_f_f.justice_interruption_rate!=0)]\n",
    "df_f_f['adv_interruption_rate'] = df_f_f.apply(lambda x: np.log(x['adv_interruption_rate']), axis=1)\n",
    "df_f_f['justice_interruption_rate'] = df_f_f.apply(lambda x: np.log(x['justice_interruption_rate']), axis=1)\n",
    "\n",
    "corr, _ = pearsonr(df_f_f.adv_interruption_rate, df_f_f.justice_interruption_rate)\n",
    "corr = np.round(corr, 2)\n",
    "#print('m-f', corr)\n",
    "axes[1][0].scatter(df_f_f.adv_interruption_rate, df_f_f.justice_interruption_rate, s=2)\n",
    "axes[1][0].set_title('male advocate, female justice', fontsize=12)\n",
    "axes[1][0].axes.text(0.8,-0.3, r'$\\rho$='+str(corr), horizontalalignment='left',\\\n",
    "                                 verticalalignment='center', transform=axes[0][0].transAxes,\\\n",
    "                                 fontsize=12)\n",
    "\n",
    "\n",
    "df_f_f = df[(df.advocate_gender=='M') & (df.justice_gender=='M')]\n",
    "df_f_f = df_f_f[['adv_interruption_rate', 'justice_interruption_rate']]\n",
    "\n",
    "df_f_f = df_f_f[(df_f_f.adv_interruption_rate!=0) & (df_f_f.justice_interruption_rate!=0)]\n",
    "df_f_f['adv_interruption_rate'] = df_f_f.apply(lambda x: np.log(x['adv_interruption_rate']), axis=1)\n",
    "df_f_f['justice_interruption_rate'] = df_f_f.apply(lambda x: np.log(x['justice_interruption_rate']), axis=1)\n",
    "\n",
    "corr, _ = pearsonr(df_f_f.adv_interruption_rate, df_f_f.justice_interruption_rate)\n",
    "corr = np.round(corr, 2)\n",
    "#print('m-m', corr)\n",
    "axes[1][1].scatter(df_f_f.adv_interruption_rate, df_f_f.justice_interruption_rate, s=2)\n",
    "axes[1][1].set_title('male advocate, male justice', fontsize=12)\n",
    "axes[1][1].axes.text(2,-0.3, r'$\\rho$='+str(corr), horizontalalignment='left',\\\n",
    "                                 verticalalignment='center', transform=axes[0][0].transAxes,\\\n",
    "                                 fontsize=12)\n",
    "\n",
    "        \n",
    "fig.text(0.5, 0.03, 'advocate interruption rate (log scale)', ha='center', fontsize=12)\n",
    "fig.text(0.05, 0.5, 'justice interruption rate (log scale)', va='center', rotation='vertical', fontsize=12)\n",
    "\n",
    "plt.savefig(\"figs/figA1-heatedness.pdf\")\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4e388c0",
   "metadata": {},
   "source": [
    "### Table 4: Corroborative mediation analyses\n",
    "\n",
    "We only use 2007 onwards in the data because speech disfluencies were not recorded consistently. \n",
    "\n",
    "Our mediators are: \n",
    "- Speech disfluencies as mediator\n",
    "- deological alignment as mediator\n",
    "- Advocate experience as mediator \n",
    "\n",
    "and we calculate the natural direct effect (NDE) and natural indirect effect (NIE) from treatment (gender of the advocate) to outcome (token-normalized interruption rate). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "07a2d95a",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_subset = df[df['case_year'] >= 2007]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "4bdc2d1a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(17801, 10)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_subset), len(set(df_subset.justice_name))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e3cdf29",
   "metadata": {},
   "source": [
    "## Corroborative analyses"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d99a9cd0",
   "metadata": {},
   "source": [
    "### Background on estimand analysis with mediation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "227fd2bb",
   "metadata": {},
   "source": [
    "#### Analysis \n",
    "\n",
    "All of these are all conditional on the chunks that the justice is speaking in \n",
    "\n",
    " Total effect \n",
    "\n",
    "$$ TE = E [Y|T=1] - E[Y|T=0] $$\n",
    "\n",
    "Pearl, Causal Inference in Statastics\n",
    "\n",
    "Eqns. 4.50 and 4.52\n",
    "\n",
    "$$ NDE = \\sum_m [E[Y|T=1, M=m] - E[Y|T=0, M=m]]\\cdot P(M=m|T=0)$$\n",
    "\n",
    "$$ NIE = \\sum_m [E[Y|T=0, M=m]] \\cdot [P(M=m|T=1)- P(M=m|T=0)]$$\n",
    "\n",
    "where \n",
    "\n",
    "$$ P(M=m|T=0) = \\frac{\\text{count\\_chunks}(M=m \\cap T=0)}{\\text{count\\_chunks}(T=0)} $$\n",
    "\n",
    "These will be per-justice values. The expected values are over all valid chunks for that justice. \n",
    "\n",
    "Our variables:  \n",
    "- $M$ (mediators) = Either (a) speech disfluency tertiles (3 values) or (b) advocate-justice ideology match (2 values)\n",
    "- $T$ (treatment) = $T=1$ if advocate female, $T=0$ if male\n",
    "- $Y$ (outcome) = interruption rate (# utts justice interrupts advocate/ # utts in a chunk)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "971d5fba",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_estimands(df_input, mediator_colm_name, treatment_colm_name, outcome_colm_name,\\\n",
    "                  num_bootstraps, justice_gender_map): \n",
    "    \n",
    "    justices = list(set(df_input['justice_name']))\n",
    "    \n",
    "    #ideology per justice\n",
    "    ideology_map = {}\n",
    "    for just in justices: \n",
    "        #print(just)\n",
    "        ideology_map[just] = df_input[df_input.justice_name==just]['justice_ideology'].iloc[0]\n",
    "        \n",
    "\n",
    "    #get estimands through mediated paths \n",
    "    df_mediated_effects = []\n",
    "    df_mediated_effects_male = []\n",
    "    df_mediated_effects_female = []\n",
    "    \n",
    "    bootsraps_for_mediated_effects = {}\n",
    "    for just in justices: \n",
    "        df_just = make_df_justice_general(just, df_input,\\\n",
    "                                        mediator_colm_name = mediator_colm_name,\\\n",
    "                                        treatment_colm_name = treatment_colm_name,\\\n",
    "                                        outcome_colm_name = outcome_colm_name)\n",
    "        \n",
    "\n",
    "        T_hat = pearls_mediation(df_just, just)\n",
    "\n",
    "        # get bootstrap std\n",
    "        bootstraps = []\n",
    "        for bs in range(num_bootstraps):\n",
    "            df_just_bs = df_just.sample(frac=1, replace=True)\n",
    "            out_bs = pearls_mediation(df_just_bs, just)\n",
    "            bootstraps.append(out_bs)\n",
    "\n",
    "        for m in ['nde', 'nie']:\n",
    "            std = get_std(bootstraps, T_hat, m)\n",
    "            T_hat['std_'+m] = std\n",
    "\n",
    "        T_hat[\"num_chunks\"] = len(df_input[df_input.justice_name==just]) \n",
    "        df_mediated_effects.append(T_hat)\n",
    "        \n",
    "        if justice_gender_map[just]=='M':\n",
    "            df_mediated_effects_male.append(T_hat)\n",
    "        if justice_gender_map[just]=='F':\n",
    "            df_mediated_effects_female.append(T_hat)\n",
    "        bootsraps_for_mediated_effects[just] = bootstraps\n",
    "        \n",
    "    df_mediated_effects = pd.DataFrame(df_mediated_effects)\n",
    "    df_mediated_effects_male = pd.DataFrame(df_mediated_effects_male)\n",
    "    df_mediated_effects_female = pd.DataFrame(df_mediated_effects_female)\n",
    "    \n",
    "    ###### NDE ######\n",
    "    \n",
    "    #average nde over all justices\n",
    "    bootsraps_for_mean_mediated_effect = []\n",
    "    for boot in range(num_bootstraps):\n",
    "        mediated_effects_over_just = []\n",
    "        for just, effects in bootsraps_for_mediated_effects.items():\n",
    "            mediated_effects_over_just.append(effects[boot]['nde'])\n",
    "        bootsraps_for_mean_mediated_effect.append(pd.DataFrame(mediated_effects_over_just, columns=['nde']))\n",
    "\n",
    "    avg_just_nde_hat = {'nde': df_mediated_effects['nde'].mean()} \n",
    "    avg_just_nde_hat['std_'+'nde'] = get_std(bootsraps_for_mean_mediated_effect, avg_just_nde_hat, 'nde')\n",
    "    \n",
    "    #average nde over male justices\n",
    "    bootsraps_for_mean_mediated_effect_male = []\n",
    "    for boot in range(num_bootstraps):\n",
    "        mediated_effects_over_just_male = []\n",
    "        for just, effects in bootsraps_for_mediated_effects.items():\n",
    "            if justice_gender_map[just]=='M':\n",
    "                mediated_effects_over_just_male.append(effects[boot]['nde'])\n",
    "        bootsraps_for_mean_mediated_effect_male.append(pd.DataFrame(mediated_effects_over_just_male,\\\n",
    "                                                                    columns=['nde']))\n",
    "\n",
    "    avg_just_male_nde_hat = {'nde': df_mediated_effects_male['nde'].mean()} \n",
    "    avg_just_male_nde_hat['std_'+'nde'] = std = get_std(bootsraps_for_mean_mediated_effect_male,\\\n",
    "                                                        avg_just_male_nde_hat,'nde')\n",
    "    \n",
    "    #average nde over female justices\n",
    "    bootsraps_for_mean_mediated_effect_female = []\n",
    "    for boot in range(num_bootstraps):\n",
    "        mediated_effects_over_just_female = []\n",
    "        for just, effects in bootsraps_for_mediated_effects.items():\n",
    "            if justice_gender_map[just]=='F':\n",
    "                mediated_effects_over_just_female.append(effects[boot]['nde'])\n",
    "        bootsraps_for_mean_mediated_effect_female.append(\\\n",
    "                                pd.DataFrame(mediated_effects_over_just_female,\\\n",
    "                                             columns=['nde']))\n",
    "\n",
    "    avg_just_female_nde_hat = {'nde': df_mediated_effects_female['nde'].mean()} \n",
    "    avg_just_female_nde_hat['std_'+'nde'] = get_std(bootsraps_for_mean_mediated_effect_female,\\\n",
    "                                                    avg_just_female_nde_hat, 'nde')\n",
    "    \n",
    "    \n",
    "    ###### NIE ######\n",
    "        \n",
    "    #average nie over all justices\n",
    "    bootsraps_for_mean_mediated_effect = []\n",
    "    for boot in range(num_bootstraps):\n",
    "        mediated_effects_over_just = []\n",
    "        for just, effects in bootsraps_for_mediated_effects.items():\n",
    "            mediated_effects_over_just.append(effects[boot]['nie'])\n",
    "        bootsraps_for_mean_mediated_effect.append(pd.DataFrame(mediated_effects_over_just, columns=['nie']))\n",
    "    \n",
    "    avg_just_nie_hat = {'nie': df_mediated_effects['nie'].mean()} \n",
    "    avg_just_nie_hat['std_'+'nie'] = get_std(bootsraps_for_mean_mediated_effect, avg_just_nie_hat, 'nie')\n",
    "    \n",
    "    #average nie over male justices\n",
    "    bootsraps_for_mean_mediated_effect_male = []\n",
    "    for boot in range(num_bootstraps):\n",
    "        mediated_effects_over_just_male = []\n",
    "        for just, effects in bootsraps_for_mediated_effects.items():\n",
    "            if justice_gender_map[just]=='M':\n",
    "                mediated_effects_over_just_male.append(effects[boot]['nie'])\n",
    "        bootsraps_for_mean_mediated_effect_male.append(pd.DataFrame(\\\n",
    "                                mediated_effects_over_just_male, columns=['nie']))\n",
    "    avg_just_male_nie_hat = {'nie': df_mediated_effects_male['nie'].mean()} \n",
    "    avg_just_male_nie_hat['std_'+'nie'] = get_std(bootsraps_for_mean_mediated_effect_male,\\\n",
    "                                                  avg_just_male_nie_hat, 'nie')\n",
    "    \n",
    "    \n",
    "    #average nie over female justices\n",
    "    bootsraps_for_mean_mediated_effect_female = []\n",
    "    for boot in range(num_bootstraps):\n",
    "        mediated_effects_over_just_female = []\n",
    "        \n",
    "        for just, effects in bootsraps_for_mediated_effects.items():\n",
    "            if justice_gender_map[just]=='F':\n",
    "                mediated_effects_over_just_female.append(effects[boot]['nie'])\n",
    "                \n",
    "        bootsraps_for_mean_mediated_effect_female.append(pd.DataFrame(mediated_effects_over_just_female,\\\n",
    "                                                         columns=['nie']))\n",
    "    avg_just_female_nie_hat = {'nie': df_mediated_effects_female['nie'].mean()} \n",
    "    avg_just_female_nie_hat['std_'+'nie'] = get_std(bootsraps_for_mean_mediated_effect_female,\\\n",
    "                                                    avg_just_female_nie_hat, 'nie')\n",
    "\n",
    "\n",
    "    return {'mediator_colm_name': mediator_colm_name,  \n",
    "            'mediated': df_mediated_effects, \n",
    "            'avg_direct_effect_over_justices':avg_just_nde_hat,\n",
    "            'avg_direct_effect_over_male_justices':avg_just_male_nde_hat,\n",
    "            'avg_direct_effect_over_female_justices':avg_just_female_nde_hat,\n",
    "            'avg_indirect_effect_over_justices':avg_just_nie_hat,\n",
    "            'avg_indirect_effect_over_male_justices':avg_just_male_nie_hat,\n",
    "            'avg_indirect_effect_over_female_justices':avg_just_female_nie_hat,\n",
    "            'ideology_map': ideology_map}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a9ea6a7c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_std(bootstraps, T_hat, metric):\n",
    "    #Using 6.16 Theorem (Normal-based Confidence Interval; 95%) in Wasserman 2004 \n",
    "    bootstrap_distribution = []\n",
    "    for b in bootstraps:\n",
    "        bootstrap_distribution.append(b[metric].mean())\n",
    "    \n",
    "    std_error = np.std(bootstrap_distribution)   \n",
    "    return std_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "ef261125",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"../raw_data/justice2gender.json\", \"r\") as f: \n",
    "    justice_gender_map = json.load(f)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e36cfe71",
   "metadata": {},
   "source": [
    "### Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "8eb2fb92",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded final df from  data/df_final.csv\n",
      "Number of rows= 64164\n",
      "----Number of chunks----\n",
      "Justices:All 17801\n",
      "Justices:Female 5980\n",
      "Justices:Male 11821\n"
     ]
    }
   ],
   "source": [
    "df = load_final_df(config) # Dataframe after filtering to filters in config.yaml \n",
    "                           # (e.g., filtering to justices with the min num chunks) \n",
    "    \n",
    "df_subset = df[df['case_year'] >= 2007]\n",
    "\n",
    "print(\"----Number of chunks----\")\n",
    "print('Justices:All', len(df_subset))\n",
    "print('Justices:Female', len(df_subset[df_subset.justice_gender=='F']))\n",
    "print('Justices:Male', len(df_subset[df_subset.justice_gender=='M']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f3e0858",
   "metadata": {},
   "source": [
    "### Mediation by speech disfluencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "954ebfab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "disfluency rate (first quantile):  14.77832512315271\n",
      "disfluency rate (second quantile):  29.41176470588235\n"
     ]
    }
   ],
   "source": [
    "### get speech disfluencies variable\n",
    "\n",
    "df_subset['adv_disfl_rate'] = 1000*df_subset['num_adv_disfl']/df_subset['num_toks_adv']\n",
    "adv_disfl_rate = df_subset['adv_disfl_rate'].to_numpy()\n",
    "\n",
    "# discretizing disfluency rates\n",
    "# let's just go with the third quantiles (tertiles) for now \n",
    "t1 = np.quantile(adv_disfl_rate, 1/3)\n",
    "t2 = np.quantile(adv_disfl_rate, 2/3)\n",
    "print('disfluency rate (first quantile): ', t1)\n",
    "print('disfluency rate (second quantile): ', t2)\n",
    "\n",
    "\n",
    "# add discretized version to the data frame \n",
    "adv_disfl_rate_tertiles = []\n",
    "for x in adv_disfl_rate: \n",
    "    if 0 <= x < t1: \n",
    "        adv_disfl_rate_tertiles.append(0)\n",
    "    elif t1 <= x < t2: \n",
    "        adv_disfl_rate_tertiles.append(1)\n",
    "    elif t2 <= x: \n",
    "        adv_disfl_rate_tertiles.append(2)\n",
    "assert len(adv_disfl_rate_tertiles) == len(adv_disfl_rate) == len(df_subset)\n",
    "df_subset['adv_disfl_rate_tertiles'] = np.array(adv_disfl_rate_tertiles).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "2d38b5a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "mediator_colm_name = 'adv_disfl_rate_tertiles'\n",
    "treatment_colm_name = 'advocate_gender'\n",
    "outcome_colm_name = 'adv_interruption_rate'\n",
    "out_sd = get_estimands(df_subset, mediator_colm_name, treatment_colm_name, outcome_colm_name,\\\n",
    "                             config['num_bootstrap_samples'], justice_gender_map)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b450adc6",
   "metadata": {},
   "source": [
    "### Mediation by ideological alignment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "2fefca98",
   "metadata": {},
   "outputs": [],
   "source": [
    "mediator_colm_name = 'ideology_matches'\n",
    "treatment_colm_name = 'advocate_gender'\n",
    "outcome_colm_name = 'adv_interruption_rate'\n",
    "out_ia = get_estimands(df_subset, mediator_colm_name, treatment_colm_name, outcome_colm_name,\\\n",
    "                             config['num_bootstrap_samples'], justice_gender_map)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "733cf37e",
   "metadata": {},
   "source": [
    "### Mediation by advocate experience"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "5ee026d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "mediator_colm_name = 'adv_experience_bin'\n",
    "treatment_colm_name = 'advocate_gender'\n",
    "outcome_colm_name = 'adv_interruption_rate'\n",
    "out_exp = get_estimands(df_subset, mediator_colm_name, treatment_colm_name, outcome_colm_name,\\\n",
    "                             config['num_bootstrap_samples'], justice_gender_map)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5689a4a1",
   "metadata": {},
   "source": [
    "### Print results for Table 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "c22351b3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Justices:All\n",
      "----NDE----\n",
      "Mediator:Speech Disfluencies 0.73 +- 0.6\n",
      "Mediator:Ideological Alignment 0.83 +- 0.69\n",
      "Mediator:Advocate Experience 0.68 +- 0.7\n",
      "----NIE----\n",
      "Mediator:Speech Disfluencies 0.07 +- 0.29\n",
      "Mediator:Ideological Alignment 0.04 +- 0.05\n",
      "Mediator:Advocate Experience -0.03 +- 0.08\n"
     ]
    }
   ],
   "source": [
    "print('Justices:All')\n",
    "print(\"----NDE----\")\n",
    "e1 = out_sd[\"avg_direct_effect_over_justices\"]['nde']\n",
    "e1_std = out_sd[\"avg_direct_effect_over_justices\"]['std_nde']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Speech Disfluencies\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_ia[\"avg_direct_effect_over_justices\"]['nde']\n",
    "e1_std = out_ia[\"avg_direct_effect_over_justices\"]['std_nde']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Ideological Alignment\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_exp[\"avg_direct_effect_over_justices\"]['nde']\n",
    "e1_std = out_exp[\"avg_direct_effect_over_justices\"]['std_nde']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Advocate Experience\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "\n",
    "print(\"----NIE----\")\n",
    "e1 = out_sd[\"avg_indirect_effect_over_justices\"]['nie']\n",
    "e1_std = out_sd[\"avg_indirect_effect_over_justices\"]['std_nie']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Speech Disfluencies\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_ia[\"avg_indirect_effect_over_justices\"]['nie']\n",
    "e1_std = out_ia[\"avg_indirect_effect_over_justices\"]['std_nie']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Ideological Alignment\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_exp[\"avg_indirect_effect_over_justices\"]['nie']\n",
    "e1_std = out_exp[\"avg_indirect_effect_over_justices\"]['std_nie']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Advocate Experience\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "af324caa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Justices:Male\n",
      "----NDE----\n",
      "Mediator:Speech Disfluencies 1.2 +- 0.78\n",
      "Mediator:Ideological Alignment 1.36 +- 0.92\n",
      "Mediator:Advocate Experience 1.18 +- 0.91\n",
      "----NIE----\n",
      "Mediator:Speech Disfluencies 0.12 +- 0.41\n",
      "Mediator:Ideological Alignment 0.06 +- 0.08\n",
      "Mediator:Advocate Experience 0.0 +- 0.11\n"
     ]
    }
   ],
   "source": [
    "print('Justices:Male')\n",
    "print(\"----NDE----\")\n",
    "e1 = out_sd[\"avg_direct_effect_over_male_justices\"]['nde']\n",
    "e1_std = out_sd[\"avg_direct_effect_over_male_justices\"]['std_nde']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Speech Disfluencies\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_ia[\"avg_direct_effect_over_male_justices\"]['nde']\n",
    "e1_std = out_ia[\"avg_direct_effect_over_male_justices\"]['std_nde']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Ideological Alignment\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_exp[\"avg_direct_effect_over_male_justices\"]['nde']\n",
    "e1_std = out_exp[\"avg_direct_effect_over_male_justices\"]['std_nde']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Advocate Experience\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "\n",
    "print(\"----NIE----\")\n",
    "e1 = out_sd[\"avg_indirect_effect_over_male_justices\"]['nie']\n",
    "e1_std = out_sd[\"avg_indirect_effect_over_male_justices\"]['std_nie']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Speech Disfluencies\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_ia[\"avg_indirect_effect_over_male_justices\"]['nie']\n",
    "e1_std = out_ia[\"avg_indirect_effect_over_male_justices\"]['std_nie']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Ideological Alignment\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_exp[\"avg_indirect_effect_over_male_justices\"]['nie']\n",
    "e1_std = out_exp[\"avg_indirect_effect_over_male_justices\"]['std_nie']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Advocate Experience\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "339ee56d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Justices:Female\n",
      "----NDE----\n",
      "Mediator:Speech Disfluencies -0.38 +- 0.84\n",
      "Mediator:Ideological Alignment -0.38 +- 0.87\n",
      "Mediator:Advocate Experience -0.48 +- 0.86\n",
      "----NIE----\n",
      "Mediator:Speech Disfluencies -0.03 +- 0.26\n",
      "Mediator:Ideological Alignment -0.01 +- 0.04\n",
      "Mediator:Advocate Experience -0.1 +- 0.1\n"
     ]
    }
   ],
   "source": [
    "print('Justices:Female')\n",
    "print(\"----NDE----\")\n",
    "e1 = out_sd[\"avg_direct_effect_over_female_justices\"]['nde']\n",
    "e1_std = out_sd[\"avg_direct_effect_over_female_justices\"]['std_nde']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Speech Disfluencies\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_ia[\"avg_direct_effect_over_female_justices\"]['nde']\n",
    "e1_std = out_ia[\"avg_direct_effect_over_female_justices\"]['std_nde']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Ideological Alignment\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_exp[\"avg_direct_effect_over_female_justices\"]['nde']\n",
    "e1_std = out_exp[\"avg_direct_effect_over_female_justices\"]['std_nde']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Advocate Experience\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "\n",
    "print(\"----NIE----\")\n",
    "e1 = out_sd[\"avg_indirect_effect_over_female_justices\"]['nie']\n",
    "e1_std = out_sd[\"avg_indirect_effect_over_female_justices\"]['std_nie']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Speech Disfluencies\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_ia[\"avg_indirect_effect_over_female_justices\"]['nie']\n",
    "e1_std = out_ia[\"avg_indirect_effect_over_female_justices\"]['std_nie']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Ideological Alignment\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))\n",
    "\n",
    "e1 = out_exp[\"avg_indirect_effect_over_female_justices\"]['nie']\n",
    "e1_std = out_exp[\"avg_indirect_effect_over_female_justices\"]['std_nie']\n",
    "e1_ci = e1_std*1.96\n",
    "print(\"Mediator:Advocate Experience\", np.round(e1, 2), \"+-\", np.round(e1_ci, 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "457f7dea",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
